累積數量管制圖應用於顧客流失之預測

Cumulative Quantity Control Chart with Application to Customer Churn Prediction

楊珮慈(Pei-Tsz Yang) ； 俞瑋婷(Wei-Ting Yu) ； 葉思佳(Ssu-Chia Yen) ； 陳思翰(Ssu-Han Chen)

《管理資訊計算》 6卷特刊2 (2017/08) Pp. 1-11

https://doi.org/10.6285/MIC.6(S2).01

摘要

本研究建構處理網路顧客流失的預測模式，以累積數量管制圖(Cumulative Quantity Control Chart，簡稱CQC管制圖)作為主要的監控工具，並結合混淆矩陣(Confusion Matrix)以找出其最適宜的參數值。該預測模式能動態地監控各個顧客登入行為是否出現惡化趨向，在監控的變數方面，間隔登入時間(Inter-login Time)可顯現出登入行為的歷史軌跡；而近時(Recency)則可掌握登入行為的近況，本研究同時將兩變數運用在CQC中。過往的預警模式大都是以靜態的方式分析顧客流失與否，本研究則使用有別以往的動態預警概念，除了視覺化(Visualization)的圖形展示之外，隨著時間的變動、資訊的更新，可以持續的在CQC上繪出新資訊，當某CQC分數超出預定的上管制界限(Upper Control Limit，簡稱UCL)，流失的警訊便會出現，代表某位顧客的登入行為出現惡化的現象。

關鍵字

顧客流失預測 ； 累積數量管制圖 ； 間隔登入時間 ； 近時

並列摘要

This study probes into the prediction model of customer churn on the Internet. This model takes the cumulative quantity control (CQC) chart as the monitoring tool in which the most appropriate parameters are found with the combination of control chart mechanism and confusion matrix. The prediction model can monitor control chart dynamically to examine whether customer's login behavior tends to deteriorate or not. In monitoring of variables, inter-login time can show the historical track of login behavior, while recency can reflect the recent login situation. This study applies two variables in CQC at the same time. Most past prediction models adopt static analysis approach to analyze customer churn. This research is using the dynamic prediction concept, which is different from that of the past. Except for visual diagram, new information will be continuously shown in the chart as time changes and information updates. When CQC score exceeds upper control limit (UCL), a warning of customer churn will appear, which represents the deterioration of certain customer's behavior.

並列關鍵字

Customer Churn Prediction ； Cumulative Quantity Control Chart ； Inter-login Time ； Recency

參考文獻

Chan, L.Y.,Xie, M.,Goh, T. N.(2000).Cumulative quantity control charts for monitoring production processes.International Journal of Production Research.38(2),397-408.

Coussement, K.,Benoit, D. F.,Van den Poel, D.(2010).Improved marketing decision making in a customer churn prediction context using generalized additive models.Expert Systems with Applications.37(3),2132-2143.

Coussement, K.,De Bock, K. W.(2013).Customer churn prediction in the online gambling industry: The beneficial effect of ensemble learning.Journal of Business Research.66(9),1629-1636.

Frederick, F. R.,Sasser, W. E.(1990).Zero defections: Quality comes to services.Harvard Business Review.68(5),105-111.

Hadiji, F.,Sifa, R.,Drachen, A.,Thurau, C.,Kersting, K.,Bauckhage, C.(2014).Predicting player churn in the wild.Computational intelligence and games (CIG), 2014 IEEE conference on.(Computational intelligence and games (CIG), 2014 IEEE conference on).